Today's Message Index:
----------------------
1. 10:06 AM - Re: Re: Voltmeter & Ammeter normal ranges (Robert L. Nuckolls, III)
2. 11:24 AM - Daily List Digest Question (jrevens)
3. 01:57 PM - Re: Dual batteries or dual alternators? (Flying Patagonia)
4. 03:46 PM - Re: Dual batteries or dual alternators? (Flying Patagonia)
5. 06:38 PM - alternator whine (Alec Myers)
6. 07:39 PM - Re: alternator whine (user9253)
7. 10:05 PM - Re: Re: Voltmeter & Ammeter normal ranges (Bob Verwey)
Message 1
INDEX | Back to Main INDEX |
NEXT | Skip to NEXT Message |
LIST | Reply to LIST Regarding this Message |
SENDER | Reply to SENDER Regarding this Message |
|
Subject: | Re: Voltmeter & Ammeter normal ranges |
At 08:32 AM 9/24/2017, you wrote:
>Exactly. Joe could you please confirm that the jpeg below is
>visible? If so it's obvious that while the voltage seems to be
>moving around a tiny bit, it never gets to the point which would
>allow the amperage shown. It's always above 14V. Also I checked some
>data with the engine shut down and while the amperage still wanders
>around it's 1 A or less (-5.5 to -6.5 A)
Digital data acquisition is fraught with
opportunity for confusion/error. Depending
on how the variable is sampled, pre-filtered
and software filtered, data displayed or plotted
can take on the appearance of instability.
Unless you're measuring the output from something
like a battery . . . or a laboratory power supply,
the stimulus being observed will have some noise
on it. This is especially true of signals gathered
from the DC power system of any vehicle.
When I'm gathering data on relatively low frequency
events, my DAS pre-filtering is usually rolled
off at 10 Hz. Then interpretation software is
designed to do a 'rolling average' on sampled values.
For a DAS set up to gather 100 samples/second, we'll
buffer 16 samples in a memory stack, take the sum
of those values then divide by 16 for an average
of that 'packet' of data. When the next sample comes
in, we add the new value, throw away the oldest value
and do the average again.
This pre/post processing of data does a nice job
of ignoring the high frequency perturbations that
are riding on top of the value of interest. The
processed data plots are nice smooth lines that
are more meaningful.
I suspect that data plots taken from most
EIS systems have little pre-processing and no
post processing . . . so the traces are understandably
'ragged'. So while the voltage traces might
be interpreted as 'unstable', it's probable
that the system is operating normally.
Just for grins, I took the 'scope out to peek at
the 14v bus in Dr. Dee's '87 GMC Sierra. With
the 'scope set to 2.5S/div H and 2V/div V, we
get the following trace when cranking the engine.
[]
Figure 1.
Note that battery voltage starts to sag from
about 12.8 volts but takes a steeper slope
as the starter contactor is energized followed
by a profound drop to 8 volts as the contactor
closes. A few milliseconds later, the voltage
rises to about 10v until the engine catches
approx 1.2S later and the alternator kicks
in . . . whereupon but bus rises to about
14.6V
[]
Figure 2.
In the next plot we see that the bus is 'steady'
at about 14.7V with the occasional perturbation
combined with some little 'spikes' with an amplitude
of 1 to 2 volts.
To get a closer look at the voltage in the
next plot, we need to turn up the 'scope sensitivity
and to to AC coupling:
[]
Figure 3.
With a vertical scale of 0.5V/div and horizontal
of 0.05S/div we can get a better look at the 'noise'.
This particular trace caught a 1.5v transient amongst
about 10 smaller artifacts.
We see that even when there are no obvious
transients, the horizontal line has 'thickness'.
Zooming in closer:
[]
Figure 4.
We find that the horizontal line is 'fuzzy' with
positive and negative going transients on the
order of 300mV peak.
Depending on where your digital data acquistion
system takese a 'snapshot', the sampled votlage
might appear to 'wobble' by as much as +/-0.5V.
Spreading out the major transients by a factor
of 20 . . .
[]
Figure 5.
Shows that they're about 400 microSeconds wide. These
plots offer a small peek into the nature of DC power
systems on vehicles. A few weeks ago, a thread on
another list server offered this image suggesting
that the full-wave, rectified 3-phase alternator
would produce wave forms something along these lines . . .
[]
Figure 6.
Theoretically true, but I've never seen so 'clean'
a picture in the wild . . .
Referring to Mil-Std-704 we learn that DC power systems
demonstrate a certain amount of noise (distortion)
under normal conditions. The document cites
the following plot for a 28v system (cut the
numbers in half for 14 volts).
[]
Figure 7.
Over the range of 1000 Hz (1 milliSecond period) and
10,000 Hz (100 uSecond), we should expect artifacts
of noise up to 0.5 VRMS (1.5 volts pk-pk) on a 14V bus.
Above and below those frequencies, the allowable/exoected
noise levels fall off. When you hook an oscilloscope to
the bus, the best you can do is sample small features
of the noise as illustrated in the first 5 figures. Further,
you'll never see the pristine plots hypothesized in Figure 6.
Emacs!
Figure 8.
The most significant attribute of a DC power system is VOLTAGE.
Referring again to Mil-STD-704 we find the plot above. Again,
cut the voltage levels in half for a 14v system.
This figure suggests that any voltage vs. time condition
below the upper trace . . . and above the lower trace . . .
is a condition to be expected for designing and operating
appliances powered from the DC system.
It's my wish that this narrative will give you some tools
for critical review of any data presented by either operating
or diagnostic instruments. Understanding characteristics
signals present on the bus -AND- limits on the measuring
instrument will go to better design, operation and diagnosis
of your ship's DC power system.
Bob . . .
Message 2
INDEX | Back to Main INDEX |
PREVIOUS | Skip to PREVIOUS Message |
NEXT | Skip to NEXT Message |
LIST | Reply to LIST Regarding this Message |
SENDER | Reply to SENDER Regarding this Message |
|
Subject: | Daily List Digest Question |
I'm pretty sure that this has been brought up before, but not sure if any answers
were provided. I receive the Daily List Digest via email, and today's digest
contained 9 messages. Following message number 9, from Charles Rogers, there
were pages and pages and pages of unreadable "code". It takes a couple of minutes
just to scroll through all of it. It seems to have been awhile since this
has happened, but sometimes it happens frequently. Is this possibly an issue
with my iPad, my ISP, etc., or does it originate with the sender? Does anyone
else experience this?
--------
John Evens
Thorp T-18 N71JE (built & flying)
Kitfox SS7 N27JE (building)
Read this topic online here:
http://forums.matronics.com/viewtopic.php?p=473367#473367
Message 3
INDEX | Back to Main INDEX |
PREVIOUS | Skip to PREVIOUS Message |
NEXT | Skip to NEXT Message |
LIST | Reply to LIST Regarding this Message |
SENDER | Reply to SENDER Regarding this Message |
|
Subject: | Re: Dual batteries or dual alternators? |
Thanks!
Do you have this diagram in better quality? It cant see very well..
Read this topic online here:
http://forums.matronics.com/viewtopic.php?p=473370#473370
Message 4
INDEX | Back to Main INDEX |
PREVIOUS | Skip to PREVIOUS Message |
NEXT | Skip to NEXT Message |
LIST | Reply to LIST Regarding this Message |
SENDER | Reply to SENDER Regarding this Message |
|
Subject: | Re: Dual batteries or dual alternators? |
Bob, or anybody else to help how to wire a dual EFII, into the Z-12 diagram...?
Thanks!
Read this topic online here:
http://forums.matronics.com/viewtopic.php?p=473371#473371
Message 5
INDEX | Back to Main INDEX |
PREVIOUS | Skip to PREVIOUS Message |
NEXT | Skip to NEXT Message |
LIST | Reply to LIST Regarding this Message |
SENDER | Reply to SENDER Regarding this Message |
|
Subject: | alternator whine |
Hi Bob and others,
My 182 has a noticeable alternator whine, about 5kHz (Im guessing) at cruise rpm.
I think its worse recently than it used to be.
Where and what should I start investigating?
Message 6
INDEX | Back to Main INDEX |
PREVIOUS | Skip to PREVIOUS Message |
NEXT | Skip to NEXT Message |
LIST | Reply to LIST Regarding this Message |
SENDER | Reply to SENDER Regarding this Message |
|
Subject: | Re: alternator whine |
Check the AC (Alternating Current) component of the alternator output. If it is
more than 1 volt, suspect a bad diode.
--------
Joe Gores
Read this topic online here:
http://forums.matronics.com/viewtopic.php?p=473379#473379
Message 7
INDEX | Back to Main INDEX |
PREVIOUS | Skip to PREVIOUS Message |
NEXT | Skip to NEXT Message |
LIST | Reply to LIST Regarding this Message |
SENDER | Reply to SENDER Regarding this Message |
|
Subject: | Re: Voltmeter & Ammeter normal ranges |
Absolutely facinating and insightful to a neophyte like myself <grin>
On 05 Oct 2017 7:17 PM, "Robert L. Nuckolls, III" <
nuckolls.bob@aeroelectric.com> wrote:
> At 08:32 AM 9/24/2017, you wrote:
>
> Exactly. Joe could you please confirm that the jpeg below is visible? If
> so it's obvious that while the voltage seems to be moving around a tiny
> bit, it never gets to the point which would allow the amperage shown. It's
> always above 14V. Also I checked some data with the engine shut down and
> while the amperage still wanders around it's 1 A or less (-5.5 to -6.5 A)
>
>
> Digital data acquisition is fraught with
> opportunity for confusion/error. Depending
> on how the variable is sampled, pre-filtered
> and software filtered, data displayed or plotted
> can take on the appearance of instability.
>
> Unless you're measuring the output from something
> like a battery . . . or a laboratory power supply,
> the stimulus being observed will have some noise
> on it. This is especially true of signals gathered
> from the DC power system of any vehicle.
>
> When I'm gathering data on relatively low frequency
> events, my DAS pre-filtering is usually rolled
> off at 10 Hz. Then interpretation software is
> designed to do a 'rolling average' on sampled values.
> For a DAS set up to gather 100 samples/second, we'll
> buffer 16 samples in a memory stack, take the sum
> of those values then divide by 16 for an average
> of that 'packet' of data. When the next sample comes
> in, we add the new value, throw away the oldest value
> and do the average again.
>
> This pre/post processing of data does a nice job
> of ignoring the high frequency perturbations that
> are riding on top of the value of interest. The
> processed data plots are nice smooth lines that
> are more meaningful.
>
> I suspect that data plots taken from most
> EIS systems have little pre-processing and no
> post processing . . . so the traces are understandably
> 'ragged'. So while the voltage traces might
> be interpreted as 'unstable', it's probable
> that the system is operating normally.
>
> Just for grins, I took the 'scope out to peek at
> the 14v bus in Dr. Dee's '87 GMC Sierra. With
> the 'scope set to 2.5S/div H and 2V/div V, we
> get the following trace when cranking the engine.
> [image: []]
>
> Figure 1.
>
> Note that battery voltage starts to sag from
> about 12.8 volts but takes a steeper slope
> as the starter contactor is energized followed
> by a profound drop to 8 volts as the contactor
> closes. A few milliseconds later, the voltage
> rises to about 10v until the engine catches
> approx 1.2S later and the alternator kicks
> in . . . whereupon but bus rises to about
> 14.6V
>
> [image: []]
> Figure 2.
>
> In the next plot we see that the bus is 'steady'
> at about 14.7V with the occasional perturbation
> combined with some little 'spikes' with an amplitude
> of 1 to 2 volts.
>
> To get a closer look at the voltage in the
> next plot, we need to turn up the 'scope sensitivity
> and to to AC coupling:
>
> [image: []]
> Figure 3.
>
> With a vertical scale of 0.5V/div and horizontal
> of 0.05S/div we can get a better look at the 'noise'.
> This particular trace caught a 1.5v transient amongst
> about 10 smaller artifacts.
>
> We see that even when there are no obvious
> transients, the horizontal line has 'thickness'.
> Zooming in closer:
>
> [image: []]
> Figure 4.
>
> We find that the horizontal line is 'fuzzy' with
> positive and negative going transients on the
> order of 300mV peak.
>
> Depending on where your digital data acquistion
> system takese a 'snapshot', the sampled votlage
> might appear to 'wobble' by as much as +/-0.5V.
>
> Spreading out the major transients by a factor
> of 20 . . .
>
>
> [image: []]
> Figure 5.
>
>
> Shows that they're about 400 microSeconds wide. These
> plots offer a small peek into the nature of DC power
> systems on vehicles. A few weeks ago, a thread on
> another list server offered this image suggesting
> that the full-wave, rectified 3-phase alternator
> would produce wave forms something along these lines . . .
>
> [image: []]
>
> Figure 6.
>
> Theoretically true, but I've never seen so 'clean'
> a picture in the wild . . .
>
> Referring to Mil-Std-704 we learn that DC power systems
> demonstrate a certain amount of noise (distortion)
> under normal conditions. The document cites
> the following plot for a 28v system (cut the
> numbers in half for 14 volts).
>
> [image: []]
>
> Figure 7.
>
> Over the range of 1000 Hz (1 milliSecond period) and
> 10,000 Hz (100 uSecond), we should expect artifacts
> of noise up to 0.5 VRMS (1.5 volts pk-pk) on a 14V bus.
>
> Above and below those frequencies, the allowable/exoected
> noise levels fall off. When you hook an oscilloscope to
> the bus, the best you can do is sample small features
> of the noise as illustrated in the first 5 figures. Further,
> you'll never see the pristine plots hypothesized in Figure 6.
>
>
> [image: Emacs!]
> Figure 8.
>
> The most significant attribute of a DC power system is VOLTAGE.
> Referring again to Mil-STD-704 we find the plot above. Again,
> cut the voltage levels in half for a 14v system.
>
> This figure suggests that any voltage vs. time condition
> below the upper trace . . . and above the lower trace . . .
> is a condition to be expected for designing and operating
> appliances powered from the DC system.
>
> It's my wish that this narrative will give you some tools
> for critical review of any data presented by either operating
> or diagnostic instruments. Understanding characteristics
> signals present on the bus -AND- limits on the measuring
> instrument will go to better design, operation and diagnosis
> of your ship's DC power system.
>
> Bob . . .
>
Other Matronics Email List Services
These Email List Services are sponsored solely by Matronics and through the generous Contributions of its members.
-- Please support this service by making your Contribution today! --
|